Article
Optimisation of pharmacy content in clinical cancer research protocols: Experience of the United Kingdom Chemotherapy and Pharmacy Advisory Service
CLINICAL TRIALS Clinical Trials 1–8 Ó The Author(s) 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1740774515569610 ctj.sagepub.com
Philip R Debruyne1,2,3, Philip J Johnson4, Lies Pottel1,3, Susanna Daniels5, Rachel Greer6, Elizabeth Hodgkinson7, Stephen Kelly8, Michelle Lycke1,3, Jens Samol9, Julie Mason10,11, Donna Kimber12, Eileen Loucaides13, Mahesh KB Parmar14, Sally Harvey15 on behalf of NIHR CRN – CPAS
Abstract Background Clarity and accuracy of the pharmacy aspects of cancer clinical trial protocols is essential. Inconsistencies and ambiguities in such protocols have the potential to delay research and jeopardise both patient safety and collection of credible data. The Chemotherapy and Pharmacy Advisory Service was established by the UK National Cancer Research Network, currently known as National Institute for Health Research Clinical Research Network, to improve the quality of pharmacy-related content in cancer clinical research protocols. This article reports the scope of Chemotherapy and Pharmacy Advisory Service, its methodology of mandated protocol review and pharmacy-related guidance initiatives and its current impact. Methods Over a 6-year period (2008–2013) since the inception of Chemotherapy and Pharmacy Advisory Service, cancer clinical trial protocols were reviewed by the service, prior to implementation at clinical trial sites. A customised Review Checklist was developed and used by a panel of experts to standardise the review process and report back queries and inconsistencies to chief investigators. Based on common queries, a Standard Protocol Template comprising specific guidance on drug-related content and a Pharmacy Manual Template were developed. In addition, a guidance framework was established to address ‘ad hoc’ pharmacy-related queries. The most common remarks made at protocol review have been summarised and categorised through retrospective analysis. In order to evaluate the impact of the service, chief investigators were asked to respond to queries made at protocol review and make appropriate changes to their protocols. Responses from chief investigators have been collated and acceptance rates determined. 1
Ageing & Cancer Research Cluster, Centre for Positive Ageing, University of Greenwich, London, UK Department of Adult Nursing & Paramedic Science, Faculty of Education & Health, University of Greenwich, London, UK 3 Cancer Centre, General Hospital Groeninge, Kortrijk, Belgium 4 Department of Molecular and Clinical Cancer Medicine, University of Liverpool, Liverpool, UK 5 Pharmacy and Medicines Management, University College London Hospitals, London, UK 6 Leeds Teaching Hospitals NHS Trust, Leeds, UK 7 Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, UK 8 Pfizer Oncology UK, Tadworth, UK 9 St George’s Hospital Healthcare NHS Trust, London, UK 10 Sandwell and West Birmingham Hospitals NHS Trust, West Midlands, UK 11 Pharmacy and Therapeutics, University of Birmingham, Birmingham, UK 12 Wessex Clinical Senate & Strategic Networks, NHS England, Southampton, UK 13 NCRI Clinical Studies Groups, London, UK 14 MRC Clinical Trials Unit, University College London, London, UK 15 NIHR CPAS, National Institute for Health Research, Clinical Research Network Cancer Coordinating Centre, Leeds, UK 2
Corresponding author: Philip R Debruyne, Ageing & Cancer Research Cluster, Centre for Positive Ageing, Department of Adult Nursing & Paramedic Science, Faculty of Education & Health, University of Greenwich, Avery Hill Campus, Grey Building, Avery Hill Rd., Eltham, London SE9 2UG, UK. Email: [email protected]; [email protected]
2
Clinical Trials
Results A total of 176 protocols were reviewed. The median number of remarks per protocol was 26, of which 20 were deemed clinically relevant and mainly concerned the drug regimen, support medication, frequency and type of monitoring and drug supply aspects. Further analysis revealed that 62% of chief investigators responded to the review. All responses were positive with an overall acceptance rate of 89% of the proposed protocol changes. Conclusion Review of pharmacy content of cancer clinical trial protocols is feasible and exposes many undetected clinically relevant issues that could hinder efficient trial conduct. Our service audit revealed that the majority of suggestions were effectively incorporated in the final protocols. The refinement of existing and development of new pharmacy-related guidance documents by Chemotherapy and Pharmacy Advisory Service might aid in better and safer clinical research. Keywords Chemotherapy, pharmacy aspects, cancer, clinical trials, quality control
Introduction Approximately 14 million adults worldwide were diagnosed with cancer in 2012, 8.2 million of them died because of the disease. In the United Kingdom, 396 per 100,000 people were confronted with a cancer diagnosis in that same year. It is estimated that the incidence of cancer will increase another 55% and 35% for men and women, respectively, between 2007 and 2030 due to growth and ageing of the population.1,2 At present, the cancer research UK website, which aims to list all cancer trials and studies recruiting in the United Kingdom, registered 1884 trials, a number that is more likely to increase rather than decrease in the near future.3 The success of a clinical trial largely depends on the quality of its protocol.4 Incompleteness, inconsistencies or errors in a protocol may impact the proper conduct of a trial with subsequent risk to patient safety and ultimately accuracy of results. As a response to the inadequacy of current research protocols, Chan et al.5 recently published guidance in the form of a checklist of recommended items to include in a clinical intervention trial protocol, although it did not include pharmacyrelated content. The latter is, however, an essential part of any clinical trial involving investigational medicinal products. Moreover, it is especially important in cancer trials, where the drugs used may be cytotoxic and/or form part of a complex treatment regimen involving multiple drugs, administered in particular orders over a number of days and frequencies. Some pharmacists had the impression that poor study design and poor pharmacy content in protocols was creating a substantial workload and hindering the set-up and running of clinical trials. Delays were being caused by issues including choice of inappropriate infusion solutions, inappropriate volumes for infusion of cytotoxic doses, enforced use of inappropriate packaging and labelling of drug supplies and complicated funding and purchasing arrangements agreed between the pharmaceutical companies and the trial organising bodies.6 In the past, oncology pharmacists also frequently reported organisational issues and
inconsistencies in clinical trials, often related to various protocol interpretations due to differences in hospital local practices. Some of these inconsistencies may have put patients at unnecessary risk of errors, increased the workload for pharmacy and nursing staff as well as jeopardised accuracy of the trial outcome. The rising number of cancer clinical trials, the more stringent national and international legislation and Good Manufacturing Practice requirements, combined with the increasing demand for pharmacy support, prompted the National Cancer Research Network (NCRN; currently known as National Institute for Health Research Clinical Research Network (NIHR CRN)) to form a standardisation committee in 2003, which evolved into the current Chemotherapy and Pharmacy Advisory Service (CPAS) by the end of 2007.7,8 It was set up to advise chief investigators, clinical trials units and clinical studies groups on the chemotherapy- and pharmacy-related content of their protocols, in order to maintain and enhance research quality and thereby aid development of high-quality research protocols. The aim was first to involve pharmacists, clinicians, nurses and pharmacy technicians at the early stages of protocol design to address problems with the protocol and underpin the ability of hospital pharmacies to support clinical trials of new and established drugs. Second, it was intended that through a transparent cycle of continuous improvement and feedback, the learning from the service would eventually mean that it was no longer required. Currently, CPAS constitutes a multidisciplinary national team of pharmacists, research nurses, haematologist and oncologists. It has a formal remit to (1) consider trials to be adopted by the NIHR CRN and review draft protocols, (2) provide support to investigators and others about medicine-related issues in oncology/haematology trials and (3) review published evidence to help standardise chemotherapy administration in clinical trials (e.g. addressing generic issues such as dosage modifications in organ dysfunction,
Debruyne et al. calculations of body surface area, modifications for obesity, etc.). CPAS reviews are mandatory for all new drug trials approved by the Cancer Research UK Clinical Trials Awards and Advisory Committee and the National Institute for Health Research Health Technology Assessment programme. The mandatory process does not involve Medical Research Council– funded trials and industry-funded studies, but they can be submitted for review on a voluntary basis. This article describes the establishment of the service, its methodology, the retrospective review of its activities for the first 6 years (from January 2008 until December 2013) and analysis of the responses from chief investigators to issues raised at review.
Methods CPAS: the organisation The CPAS core comprises four ex officio members, or non-reviewers, namely, the chair, the representatives of, respectively, the NIHR CRN and National Cancer Research Institute (NCRI) clinical studies group secretariat and an NIHR CRN pharmacy advisor. The latter serves as the main point of contact and liaison between researchers and CPAS and coordinates CPAS activities and the advisory service as a whole. At present, CPAS membership includes 50 Panel and 15 Committee members, consisting of 37 pharmacists, 13 clinicians, 5 research nurses, 5 pharmacy technicians, 1 clinical trials unit manager and the four core Committee members. All CPAS non-core members are responsible for protocol reviewing and other protocol- or pharmacy-related
3 queries. The Committee members also fulfil a strategic decision-making role.
Review of draft protocols A Review Checklist was developed based on a literature review on clinical drug research guidelines, medication errors in cancer chemotherapy and common pharmacyrelated issues. The checklist was developed to standardise the conduct of reviews by the CPAS panel of experts and was used to report back queries and inconsistencies to chief investigators. The current checklist consists of 12 sections with a total of 119 questions (Supplementary Online Appendix 1).9 The sections include (1) regimen (nomenclature, etc.), (2) support medication, (3) dose calculation, (4) inclusion/exclusion criteria, (5) randomisation, (6) monitoring, (7) dose modification/delay, (8) drug information/concomitant medication, (9) drug administration, (10) drug supplies, (11) drug accountability/drug returns and (12) general. The first 11 categories are considered to be of clinical significance, while the 12th category pools comments that are related to the general formatting and grammar, trial administration or financial issues. Reviewers are encouraged to add any relevant comments not covered by the standard checklist. A draft protocol can be submitted for review at any point after funding approval, once the drug treatment section of the protocol is near completion, but no later than 2 months prior to multi-centre research ethics committee submission (Figure 1). Initially, a minimum of three reviewers, of which at least one was an oncology pharmacist, reviewed each protocol in parallel.
Figure 1. Diagram showing the typical protocol development timelines. CTAAC: Clinical Trials Awards and Advisory Committee; FSC: Feasibility Study Committee; HTA: Health Technology Assessment Programme; MREC: Multi-centre Research Ethics Committee.
4 Currently, the number of reviewers for each protocol averages five, namely, one clinician, a research nurse and three pharmacists with different subspecialties. Reviewers are given 2–3 weeks turnaround time. The pharmacy advisor receives, collates and edits all final reviews into one anonymised document and returns it to the respective chief investigator within 4–6 weeks of submission. Whether or not the recommendations are incorporated into the final protocol remains at the chief investigator’s discretion.
Clinical Trials different viewpoints and suggestions are discussed, and a best practice approach is agreed. All queries and responses are stored for reference. This initiative has highlighted frequently posed questions which, for example, in March 2012 led to publication of an online investigational medicinal product statement12 defining which drugs in a clinical trial protocol, are classified as investigational medicinal products and which are noninvestigational medicinal products.
Retrospective analysis of protocol reviews Support to investigators about medicine- and pharmacy-related content In addition to its review activity, CPAS provides pharmacy-related support, either by direct contact or through guidance documents to assist protocol writing. First, a Standard Protocol Template, detailing specific ‘Guidance on the drug-related content of clinical trial protocols’ was created based on the aforementioned Review Checklist and finalised in 2008.10 It is subdivided into eight sections: (1) trial procedures, (2) treatment of patients, (3) trial drugs, (4) glossary of formulae, (5) suggested capecitabine dose banding table, (6) manipulation of investigational medicinal products in the pharmacy, (7) labelling of investigational medicinal products and (8) note on oral anticancer therapy. It provides useful examples of phrases that could be incorporated in a protocol. A copy of the document can be found as the Supplementary Online Appendix 2 or on the website of the NIHR CRN.10 Second, in June 2009, a Pharmacy Manual Template was created for guidance to investigators with the content of pharmacy manuals for clinical trials.11 It contains the following sections: (1) contact details of sponsor, (2) trial synopsis, (3) study medication, (4) randomisation, (5) prescribing, (6) dispensing, (7) accountability forms, (8) patient returns, (9) destruction, (10) hazards and (11) forms/templates. All documents are available online for use by others via the NIHR CRN website or in the Supplementary Online Appendix 3. Finally, as a unique group of national ‘experts’, CPAS, are available to answer ‘ad hoc’ questions addressing pharmacy-related queries, or requests for advice, in relation to National Institute for Health Research portfolio studies. The queries range from specific study-related questions to general trial-related questions (e.g. use of electronic prescribing system, patient randomisation faxes, transportation of refrigerated Investigational Medicinal Product (IMPs) between hospital and satellite unit) and can be submitted to the pharmacy advisor. The pharmacy advisor considers the incoming queries and contacts relevant members of the CPAS panel for comment and advice. The comments are then collated into a final anonymised response based on consensus opinion. Where opinion varies, the
All protocol review reports returned to the chief investigators for the 6-year period, between 1 January 2008 and 31 December 2013, were analysed retrospectively. A detailed list of all of the 176 protocols reviewed can be found on the CPAS page of the NIHR CRN website and in the Supplementary Online Appendix 4.13 Trial characteristics were collected, and remarks that were retained in the final review report were summarised and categorised according to the 12 subsections of the Review Checklist described earlier.
Evaluation survey At the time of receiving the final collated review for their protocol, a request was made to all chief investigators (i.e. those submitting their draft protocol for review) to provide feedback. They were asked to state whether or not they agreed with the issues raised at review and provide confirmation of changes made to their protocol as a result. A service evaluation audit was conducted to check chief investigator response rates and acceptance rates (% of remarks raised at review that were accepted and reflected in changes to the protocol) of proposed changes. This was done for all protocols reviewed in this 6-year period.
Statistical analysis Descriptive statistics were performed to present trial characteristics, frequency and type of remarks and response and acceptance rates. A response rate of 60% is considered as an acceptable level of response rate to surveys.14 All analyses were conducted using MicrosoftÒ Excel 2011 (Microsoft, Inc., Redmond, WA) and IBM SPSSÒ v.19 (SPSS, Inc., Chicago, IL) software.
Results Protocols reviewed by CPAS and findings Trial characteristics of the 176 protocols that were reviewed are described in Table 1. The median number of protocols per year reviewed was 27 (range: 25–42) and appears to be stable over the years. Of these, 4%
Debruyne et al.
5
Table 1. Characteristics of the clinical trials reviewed between 2008 and 2013. Characteristic
Year of review 2008 2009 2010 2011 2012 2013 Type of trial Phase I Phase I/II Phase II Phase II/III Phase III Phase IV Other Clinical specialty group Biomarkers and imaging Bladder (including penile) Brain Breast Children’s cancer and leukaemia Colorectal Gynaecological Haematological oncology Head and neck Lung Lymphoma Melanoma Palliative and supportive care Primary care Prostate Psychosocial oncology Renal (including adrenal) Sarcoma Teenage and young adults Testis Upper gastrointestinal (including pancreas and liver) Combined Type of investigational therapy Cytotoxic chemotherapy Hormonal therapy Molecular targeted therapy Immunotherapy Combination of drugs Combination with radiotherapy Other
Studies (N = 176) No.
%
29 26 28 42 25 26
16 15 16 24 14 15
7 19 90 13 44 1 2
4 11 51 7 25 1 1
0 9 11 16 3 14 16 26 6 10 13 7 1 0 8 0 7 8 0 2 10
0 5 6 9 2 9 9 15 3 6 7 4 1 0 5 0 4 5 0 1 6
9
5
43 3 42 1 63 15 9
24 2 24 1 36 9 5
were phase I trials, 51% were phase II trials, 25% were phase III trials, 1% were phase IV trials and 18% were combined phase I/II or II/III trials. The average time between submission of the protocol to CPAS and the return of the collated review was 35 days. This figure fluctuates between 21 days (and in one or two exceptional cases an excess of 80 days), depending on how busy CPAS service becomes at any one time. The disease categories subdivided according to the respective clinical specialty groups are listed in Table 1.
The included experimental treatment modalities concerned mainly cytotoxic chemotherapy (24%), molecular targeted therapy (24%) or a combination of drugs (36%). About 9% concerned studies of drug combinations with radiotherapy, 2% anti-hormonal treatment and 1% immunotherapy. The review findings according to the 12 categories in the Protocol Review Checklist and the relative frequencies that each issue has arisen in the respective protocols are summarised in Figure 2(a) and (b). The median number of remarks per protocol was 26, of which 20 were deemed clinically relevant. In our experience, the nature of the comments raised by reviewers fell into two broad categories: missing information and insufficient clarity of the information or guidance provided. The majority of clinically relevant remarks concerned the regimen (median [Q1, Q3]; 3 [1, 6]), support medication (2 [1, 4]), monitoring (2 [1, 3]) and drug supply aspects (2 [1, 4]) and were the same over the years (Figure 2(a)). Some typical examples are listed in Table 2.
Service evaluation survey A service evaluation survey was systematically sent to the respective chief investigators, of which 62% responded. All responses were positive (qualitative responses; data not shown), which were reflected in a median overall acceptance rate of 89% of the CPAS proposed protocol changes. Response rates and acceptance rates seemed to remain stable over the respective years (data not shown).
Discussion Cancer prevalence is high and will only increase in the near future. The quality of research protocols, in particular the pharmacy-related content, is of utmost importance to ensure cost-efficient, timely and safe research. To the best of our knowledge, an initiative like CPAS is unique, and such activities that aim to support chief investigators improve the pharmacy-related quality of cancer clinical trial protocols have not been published earlier. This article describes the development of CPAS and reviews its activities from 2008 until 2013. The aim of CPAS is to raise awareness of the type, frequency and consequences of research protocol inadequacies and to provide support to investigators either directly or in the form of guidance documents to assist the development of high-quality clinical trial protocols. CPAS was established in response to a general perception by our research community that pharmacy departments were a barrier to starting and running clinical trials especially those involving complex chemotherapy regimens. It was not the intention of pharmacy departments to hinder research, and within our pharmacy community, it was recognised that the workload generated by inadequate pharmacy information
6
Clinical Trials
Figure 2. (a) Number of remarks per review category per protocol, presented as boxplots, graphically displaying median, interquartile range and minimum and maximum data values and (b) relative frequency of remarks per review category presented as a pie chart.
within protocols often caused resource issues and subsequent delays to research.6 A number of issues were raised that hindered trial implementation at the local level and which caused delays and problems in areas such as dose adjustments, dose capping, missing pharmacy information, supply of drugs and safe administration of chemotherapy.6 Local practice often differs between hospitals causing inconsistent or wrong interpretation, which can negatively affect trial conduct and consequently data accuracy. Moreover, any missing or unclear protocol information has the potential to adversely affect patient safety. The remit of CPAS is, therefore, to resolve these issues at the draft protocol stage and achieve consistency across the clinical trial portfolio hosted by the NIHR CRN. Qualitative evaluation of the CPAS review process, through the service evaluation survey, presented positive feedback. Indeed, the majority of investigators provided a written response to the final collated review,
reporting that it was a helpful and constructive process that, in their opinion, reduced the number of protocol amendments that were required during the trial. The effectiveness of the service was also demonstrated by the remarkably high acceptance rate of the remarks raised by CPAS. The chief investigators accepted almost 9 out of 10 proposed amendments, which underscore the relevance of CPAS. While CPAS activities have been appreciated by the chief investigators of cancer clinical drug trials, the cost–benefit of this timeextensive review process is not verifiable at present. Future research could, therefore, prospectively examine whether an upfront pharmacy review process reduces the number of required pharmacy-related amendments, treatment-related protocol violations or the percentage of treatment-related hospitalisations or deaths. Moreover, future work to inspect the characteristics of clinical drug trials where the chief investigators did not respond to review comments or incorporate suggested
Debruyne et al.
7
Table 2. Examples of common relevant findings. Regimen No information on which drugs are IMPs and therefore which need accountability Information missing on infusion times, stability of a product, diluents, use of non-PVC infusion bags and giving sets Information missing on what to do if a patient vomits following a dose or misses a dose Use of brand names instead of generic Use of drug names that are not used in the United Kingdom, for example, acetaminophen instead of paracetamol Information copied from previous protocol resulting in incorrect information being stated Support medication No advice given for supportive medicines, for example, pre-meds, anti-emetics, hydration and so on 2 mg/L magnesium sulphate and 20 mmol post-hydration bags insisted on by protocol (2 mg/L = 0.008 mmol magnesium per litre) Dose calculation No information on frequency of re-calculation of BSA or formula to use, re-calculation of GFR and method of GFR calculation No reference to dose banding Dose banding table of chemotherapy with doses expected to be measured to two decimal places Monitoring Information missing on haematological and biochemical monitoring Different cut-offs specified in different areas of the protocol for discontinuing a drug due to renal impairment Screening investigations specified to be carried out within 14 days of treatment. Schedule of events table did not make this clear Drug administration Incorrect description of drug administration Different drug preparation available, incorrect choice made for route of administration required Proposed drug administration is not feasible in the specific study population GFR: glomerular filtration rate; PVC: poly(vinyl chloride); IMP: investigational medicinal product; BSA: body surface area.
changes could identify areas of further guidance or educational resource that CPAS could provide. Retrospective analysis did, at present, not reveal any differences in trial phase, clinical specialty group or type of investigational therapy between cooperative and non-cooperative chief investigators (data not shown). The latter might have been reluctant to wait for CPAS feedback since it is recognised that the turnaround time for CPAS review and final collation could be seen as an addition of a substantial amount of time to an already lengthy protocol development period. As such, CPAS suggests that draft protocols are submitted as soon as possible, and the review is conducted in parallel with other protocol development processes. Currently, the number of reviewers determined to scrutinise a draft protocol for CPAS is chosen to give a range of experience, specialities and views from different regions in the United Kingdom. It is possible that future refinements to the Review Checklist and elaboration on the detail of guidance documents already provided might reduce the number of reviewers required and/or the time period of the review process. Our systematic retrospective analysis indicated that a median of 26 remarks were suggested per protocol and that the majority of remarks addressed by the reviewers were deemed clinically relevant. To our knowledge, this is the first time that such information, based on the use of a customised Review Checklist, is available. A formal request for more information about the protocol review process was sent to several cancer cooperative networks, however, only reciprocated by the National Cancer Institute and the Swiss Group for Clinical Cancer Research. Industry-funded trials or
trials executed within a large cooperative group often provide in-house protocol review by scientific disease-specific committees, not always incorporating a pharmacist. However, there is only limited or no transparency regarding the protocol review process, and thus, it serves to no benefit to external (academic) investigators. Our results indicate that the most frequently observed inconsistencies concerned the drug regimen, support medication, monitoring and drug supply aspects, thus highlighting the importance of collaboration between the oncology physician and clinical pharmacist at the time of protocol design. CPAS protocol review included trial protocols of all phases, an extensive number of clinical specialty groups and a wide variety of cancer treatments. Results might, however, not be extrapolable to commercially funded trials. Future prospective research, addressing trials sponsored by industry and different funding agencies, might be useful to explore the potential for differences in protocol quality. Over the years that CPAS has been operating, the number of review remarks per protocol appears to have remained stable. This fact could indicate that CPAS is unlikely to become redundant in the future. On the other hand, it could also suggest that CPAS guidance documents and checklists are not well known by investigators and the CPAS advisory function not well utilised. There are many factors involved in determining why an organisation does not learn from previous experience, and these could be explored by CPAS to maximise their effectiveness. Sensitisation of other cooperative groups in oncology and other medical subspecialties might also be required to ensure a more
8 widespread implementation of CPAS knowledge in the future. Continuous service evaluation audits and evaluation of the activities will no doubt lead to further service improvements and adaption of CPAS tools. Moreover, with growth, the number of activities performed by CPAS may also further expand to include advice on practical clinical trial issues relevant to pharmacists, such as dose banding, chemotherapy stability and compatibility issues and the incidence and avoidance of chemotherapy medication errors. In conclusion, we have described the development and activities of CPAS. Systematic analysis of mandated reviews of pharmacy-related aspects of cancer clinical trial protocols proved to be useful and improved the quality of the clinical trials hosted by the NIHR CRN. Moreover, with the refinement of previously published CPAS guidance, development of new CPAS guidance and lessons learned from the review process itself, we hope that this article will lead to a more widespread implementation of our knowledge by other academic groups and better, safer clinical research. Acknowledgements We would like to acknowledge the voluntary service of CPAS committee and panel members, the cooperation of clinical trial units and chief investigators managing the academic studies, staff across the Clinical Research Networks, the Clinical Studies’ Groups, Department of Health (England), the Welsh Assembly Government, the Scottish Government, the Northern Ireland Department of Health and Social Services, Cancer Research UK, Medical Research Council and members of the National Cancer Research Institute. We would also like to thank Mrs Liz Gardner for her careful reading of this manuscript.
Declaration of conflicting interests The authors have declared no conflicts of interest.
Funding Currently, CPAS is funded by NIHR CRN Cancer (formerly known as the National Cancer Research Network, or NCRN). The national pharmacy advisor post is also funded by NIHR CRN Cancer, hosted by the University of Leeds.
References 1. Ferlay J, Soerjomataram II, Dikshit R, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer 2015; 136: E359–E386. 2. Mistry M, Parkin DM, Ahmad AS, et al. Cancer incidence in the United Kingdom: projections to the year 2030. Br J Cancer 2011; 105: 1795–1803.
Clinical Trials 3. Cancer Research UK. Clinical trials and research, http:// www.cancerresearchuk.org (accessed September 2014). 4. Debruyne PR, Knott VE and Pattison NA. A call for rigorous research and transparent study reporting. Eur J Cancer Care 2013; 22: 421–422. 5. Chan AW, Tetzlaff JM, Altman DG, et al. SPIRIT 2013 statement: defining standard protocol items for clinical trials. Ann Int Med 2013; 158: 200–207. 6. Blake D, Root T, Summerhayes M, et al. The NHS cancer plan and the pharmacy contribution to cancer care, 2001, http://www.bopawebsite.org/contentimages/publi cations/OncologyPharmacy.pdf 7. Stead M, Cameron D, Lester N, et al. Strengthening clinical cancer research in the United Kingdom. Br J Cancer 2011; 104: 1529–1534. 8. NIHR CRN Cancer. National Cancer Research Network (NCRN), http://www.crn.nihr.ac.uk/cancer/about-cancerresearch/chemotherapy-and-pharmacy-advisory-service (2013, accessed March 2014). 9. NIHR CRN Cancer. Chemotherapy and Pharmacy Advisory Service (CPAS) on behalf of the National Institute for Health Research Clinical Research Network (NIHR CRN) Cancer. Protocol review guidelines, http:// www.crn.nihr.ac.uk/cancer/about-cancer-research/chemotherapy-and-pharmacy-advisory-service (2013, accessed March 2014). 10. NIHR CRN Cancer. Chemotherapy and Pharmacy Advisory Service (CPAS) on behalf of the National Institute for Health Research Clinical Research Network (NIHR CRN) Cancer. Guidance on the drug-related content of clinical trial protocols, http://www.crn.nihr.ac.uk/cancer/ about-cancer-research/chemotherapy-and-pharmacyadvisory-service (2008, accessed March 2014). 11. NIHR CRN Cancer. Chemotherapy and Pharmacy Advisory Service (CPAS) on behalf of the National Institute for Health Research Clinical Research Network (NIHR CRN) Cancer. Guidance on the content of a pharmacy manual to support clinical trial protocols, http://www.crn.nihr.ac.uk/cancer/about-cancer-research/ chemotherapy-and-pharmacy-advisory-service (2012, accessed March 2014). 12. NIHR CRN Cancer. Chemotherapy and Pharmacy Advisory Service (CPAS) on behalf of the National Institute for Health Research Clinical Research Network (NIHR CRN) Cancer. Investigational Medicinal Product statement, http://www.crn.nihr.ac.uk/cancer/about-cancerresearch/chemotherapy-and-pharmacy-advisory-service (2011, accessed March 2014). 13. NIHR CRN Cancer. Chemotherapy and Pharmacy Advisory Service (CPAS) on behalf of the National Institute for Health Research Clinical Research Network (NIHR CRN) Cancer. List of protocols reviewed by CPAS to date, http://www.crn.nihr.ac.uk/cancer/about-cancerresearch/chemotherapy-and-pharmacy-advisory-service (2014, accessed March 2014). 14. Johnson TP and Wislar JS. Response rates and nonresponse errors in surveys. JAMA 2012; 307: 1805–1806.
Optimisation of pharmacy content in clinical cancer research protocols: Experience of the United Kingdom Chemotherapy and Pharmacy Advisory Service
CLINICAL TRIALS Clinical Trials 1–8 Ó The Author(s) 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1740774515569610 ctj.sagepub.com
Philip R Debruyne1,2,3, Philip J Johnson4, Lies Pottel1,3, Susanna Daniels5, Rachel Greer6, Elizabeth Hodgkinson7, Stephen Kelly8, Michelle Lycke1,3, Jens Samol9, Julie Mason10,11, Donna Kimber12, Eileen Loucaides13, Mahesh KB Parmar14, Sally Harvey15 on behalf of NIHR CRN – CPAS
Abstract Background Clarity and accuracy of the pharmacy aspects of cancer clinical trial protocols is essential. Inconsistencies and ambiguities in such protocols have the potential to delay research and jeopardise both patient safety and collection of credible data. The Chemotherapy and Pharmacy Advisory Service was established by the UK National Cancer Research Network, currently known as National Institute for Health Research Clinical Research Network, to improve the quality of pharmacy-related content in cancer clinical research protocols. This article reports the scope of Chemotherapy and Pharmacy Advisory Service, its methodology of mandated protocol review and pharmacy-related guidance initiatives and its current impact. Methods Over a 6-year period (2008–2013) since the inception of Chemotherapy and Pharmacy Advisory Service, cancer clinical trial protocols were reviewed by the service, prior to implementation at clinical trial sites. A customised Review Checklist was developed and used by a panel of experts to standardise the review process and report back queries and inconsistencies to chief investigators. Based on common queries, a Standard Protocol Template comprising specific guidance on drug-related content and a Pharmacy Manual Template were developed. In addition, a guidance framework was established to address ‘ad hoc’ pharmacy-related queries. The most common remarks made at protocol review have been summarised and categorised through retrospective analysis. In order to evaluate the impact of the service, chief investigators were asked to respond to queries made at protocol review and make appropriate changes to their protocols. Responses from chief investigators have been collated and acceptance rates determined. 1
Ageing & Cancer Research Cluster, Centre for Positive Ageing, University of Greenwich, London, UK Department of Adult Nursing & Paramedic Science, Faculty of Education & Health, University of Greenwich, London, UK 3 Cancer Centre, General Hospital Groeninge, Kortrijk, Belgium 4 Department of Molecular and Clinical Cancer Medicine, University of Liverpool, Liverpool, UK 5 Pharmacy and Medicines Management, University College London Hospitals, London, UK 6 Leeds Teaching Hospitals NHS Trust, Leeds, UK 7 Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, UK 8 Pfizer Oncology UK, Tadworth, UK 9 St George’s Hospital Healthcare NHS Trust, London, UK 10 Sandwell and West Birmingham Hospitals NHS Trust, West Midlands, UK 11 Pharmacy and Therapeutics, University of Birmingham, Birmingham, UK 12 Wessex Clinical Senate & Strategic Networks, NHS England, Southampton, UK 13 NCRI Clinical Studies Groups, London, UK 14 MRC Clinical Trials Unit, University College London, London, UK 15 NIHR CPAS, National Institute for Health Research, Clinical Research Network Cancer Coordinating Centre, Leeds, UK 2
Corresponding author: Philip R Debruyne, Ageing & Cancer Research Cluster, Centre for Positive Ageing, Department of Adult Nursing & Paramedic Science, Faculty of Education & Health, University of Greenwich, Avery Hill Campus, Grey Building, Avery Hill Rd., Eltham, London SE9 2UG, UK. Email: [email protected]; [email protected]
2
Clinical Trials
Results A total of 176 protocols were reviewed. The median number of remarks per protocol was 26, of which 20 were deemed clinically relevant and mainly concerned the drug regimen, support medication, frequency and type of monitoring and drug supply aspects. Further analysis revealed that 62% of chief investigators responded to the review. All responses were positive with an overall acceptance rate of 89% of the proposed protocol changes. Conclusion Review of pharmacy content of cancer clinical trial protocols is feasible and exposes many undetected clinically relevant issues that could hinder efficient trial conduct. Our service audit revealed that the majority of suggestions were effectively incorporated in the final protocols. The refinement of existing and development of new pharmacy-related guidance documents by Chemotherapy and Pharmacy Advisory Service might aid in better and safer clinical research. Keywords Chemotherapy, pharmacy aspects, cancer, clinical trials, quality control
Introduction Approximately 14 million adults worldwide were diagnosed with cancer in 2012, 8.2 million of them died because of the disease. In the United Kingdom, 396 per 100,000 people were confronted with a cancer diagnosis in that same year. It is estimated that the incidence of cancer will increase another 55% and 35% for men and women, respectively, between 2007 and 2030 due to growth and ageing of the population.1,2 At present, the cancer research UK website, which aims to list all cancer trials and studies recruiting in the United Kingdom, registered 1884 trials, a number that is more likely to increase rather than decrease in the near future.3 The success of a clinical trial largely depends on the quality of its protocol.4 Incompleteness, inconsistencies or errors in a protocol may impact the proper conduct of a trial with subsequent risk to patient safety and ultimately accuracy of results. As a response to the inadequacy of current research protocols, Chan et al.5 recently published guidance in the form of a checklist of recommended items to include in a clinical intervention trial protocol, although it did not include pharmacyrelated content. The latter is, however, an essential part of any clinical trial involving investigational medicinal products. Moreover, it is especially important in cancer trials, where the drugs used may be cytotoxic and/or form part of a complex treatment regimen involving multiple drugs, administered in particular orders over a number of days and frequencies. Some pharmacists had the impression that poor study design and poor pharmacy content in protocols was creating a substantial workload and hindering the set-up and running of clinical trials. Delays were being caused by issues including choice of inappropriate infusion solutions, inappropriate volumes for infusion of cytotoxic doses, enforced use of inappropriate packaging and labelling of drug supplies and complicated funding and purchasing arrangements agreed between the pharmaceutical companies and the trial organising bodies.6 In the past, oncology pharmacists also frequently reported organisational issues and
inconsistencies in clinical trials, often related to various protocol interpretations due to differences in hospital local practices. Some of these inconsistencies may have put patients at unnecessary risk of errors, increased the workload for pharmacy and nursing staff as well as jeopardised accuracy of the trial outcome. The rising number of cancer clinical trials, the more stringent national and international legislation and Good Manufacturing Practice requirements, combined with the increasing demand for pharmacy support, prompted the National Cancer Research Network (NCRN; currently known as National Institute for Health Research Clinical Research Network (NIHR CRN)) to form a standardisation committee in 2003, which evolved into the current Chemotherapy and Pharmacy Advisory Service (CPAS) by the end of 2007.7,8 It was set up to advise chief investigators, clinical trials units and clinical studies groups on the chemotherapy- and pharmacy-related content of their protocols, in order to maintain and enhance research quality and thereby aid development of high-quality research protocols. The aim was first to involve pharmacists, clinicians, nurses and pharmacy technicians at the early stages of protocol design to address problems with the protocol and underpin the ability of hospital pharmacies to support clinical trials of new and established drugs. Second, it was intended that through a transparent cycle of continuous improvement and feedback, the learning from the service would eventually mean that it was no longer required. Currently, CPAS constitutes a multidisciplinary national team of pharmacists, research nurses, haematologist and oncologists. It has a formal remit to (1) consider trials to be adopted by the NIHR CRN and review draft protocols, (2) provide support to investigators and others about medicine-related issues in oncology/haematology trials and (3) review published evidence to help standardise chemotherapy administration in clinical trials (e.g. addressing generic issues such as dosage modifications in organ dysfunction,
Debruyne et al. calculations of body surface area, modifications for obesity, etc.). CPAS reviews are mandatory for all new drug trials approved by the Cancer Research UK Clinical Trials Awards and Advisory Committee and the National Institute for Health Research Health Technology Assessment programme. The mandatory process does not involve Medical Research Council– funded trials and industry-funded studies, but they can be submitted for review on a voluntary basis. This article describes the establishment of the service, its methodology, the retrospective review of its activities for the first 6 years (from January 2008 until December 2013) and analysis of the responses from chief investigators to issues raised at review.
Methods CPAS: the organisation The CPAS core comprises four ex officio members, or non-reviewers, namely, the chair, the representatives of, respectively, the NIHR CRN and National Cancer Research Institute (NCRI) clinical studies group secretariat and an NIHR CRN pharmacy advisor. The latter serves as the main point of contact and liaison between researchers and CPAS and coordinates CPAS activities and the advisory service as a whole. At present, CPAS membership includes 50 Panel and 15 Committee members, consisting of 37 pharmacists, 13 clinicians, 5 research nurses, 5 pharmacy technicians, 1 clinical trials unit manager and the four core Committee members. All CPAS non-core members are responsible for protocol reviewing and other protocol- or pharmacy-related
3 queries. The Committee members also fulfil a strategic decision-making role.
Review of draft protocols A Review Checklist was developed based on a literature review on clinical drug research guidelines, medication errors in cancer chemotherapy and common pharmacyrelated issues. The checklist was developed to standardise the conduct of reviews by the CPAS panel of experts and was used to report back queries and inconsistencies to chief investigators. The current checklist consists of 12 sections with a total of 119 questions (Supplementary Online Appendix 1).9 The sections include (1) regimen (nomenclature, etc.), (2) support medication, (3) dose calculation, (4) inclusion/exclusion criteria, (5) randomisation, (6) monitoring, (7) dose modification/delay, (8) drug information/concomitant medication, (9) drug administration, (10) drug supplies, (11) drug accountability/drug returns and (12) general. The first 11 categories are considered to be of clinical significance, while the 12th category pools comments that are related to the general formatting and grammar, trial administration or financial issues. Reviewers are encouraged to add any relevant comments not covered by the standard checklist. A draft protocol can be submitted for review at any point after funding approval, once the drug treatment section of the protocol is near completion, but no later than 2 months prior to multi-centre research ethics committee submission (Figure 1). Initially, a minimum of three reviewers, of which at least one was an oncology pharmacist, reviewed each protocol in parallel.
Figure 1. Diagram showing the typical protocol development timelines. CTAAC: Clinical Trials Awards and Advisory Committee; FSC: Feasibility Study Committee; HTA: Health Technology Assessment Programme; MREC: Multi-centre Research Ethics Committee.
4 Currently, the number of reviewers for each protocol averages five, namely, one clinician, a research nurse and three pharmacists with different subspecialties. Reviewers are given 2–3 weeks turnaround time. The pharmacy advisor receives, collates and edits all final reviews into one anonymised document and returns it to the respective chief investigator within 4–6 weeks of submission. Whether or not the recommendations are incorporated into the final protocol remains at the chief investigator’s discretion.
Clinical Trials different viewpoints and suggestions are discussed, and a best practice approach is agreed. All queries and responses are stored for reference. This initiative has highlighted frequently posed questions which, for example, in March 2012 led to publication of an online investigational medicinal product statement12 defining which drugs in a clinical trial protocol, are classified as investigational medicinal products and which are noninvestigational medicinal products.
Retrospective analysis of protocol reviews Support to investigators about medicine- and pharmacy-related content In addition to its review activity, CPAS provides pharmacy-related support, either by direct contact or through guidance documents to assist protocol writing. First, a Standard Protocol Template, detailing specific ‘Guidance on the drug-related content of clinical trial protocols’ was created based on the aforementioned Review Checklist and finalised in 2008.10 It is subdivided into eight sections: (1) trial procedures, (2) treatment of patients, (3) trial drugs, (4) glossary of formulae, (5) suggested capecitabine dose banding table, (6) manipulation of investigational medicinal products in the pharmacy, (7) labelling of investigational medicinal products and (8) note on oral anticancer therapy. It provides useful examples of phrases that could be incorporated in a protocol. A copy of the document can be found as the Supplementary Online Appendix 2 or on the website of the NIHR CRN.10 Second, in June 2009, a Pharmacy Manual Template was created for guidance to investigators with the content of pharmacy manuals for clinical trials.11 It contains the following sections: (1) contact details of sponsor, (2) trial synopsis, (3) study medication, (4) randomisation, (5) prescribing, (6) dispensing, (7) accountability forms, (8) patient returns, (9) destruction, (10) hazards and (11) forms/templates. All documents are available online for use by others via the NIHR CRN website or in the Supplementary Online Appendix 3. Finally, as a unique group of national ‘experts’, CPAS, are available to answer ‘ad hoc’ questions addressing pharmacy-related queries, or requests for advice, in relation to National Institute for Health Research portfolio studies. The queries range from specific study-related questions to general trial-related questions (e.g. use of electronic prescribing system, patient randomisation faxes, transportation of refrigerated Investigational Medicinal Product (IMPs) between hospital and satellite unit) and can be submitted to the pharmacy advisor. The pharmacy advisor considers the incoming queries and contacts relevant members of the CPAS panel for comment and advice. The comments are then collated into a final anonymised response based on consensus opinion. Where opinion varies, the
All protocol review reports returned to the chief investigators for the 6-year period, between 1 January 2008 and 31 December 2013, were analysed retrospectively. A detailed list of all of the 176 protocols reviewed can be found on the CPAS page of the NIHR CRN website and in the Supplementary Online Appendix 4.13 Trial characteristics were collected, and remarks that were retained in the final review report were summarised and categorised according to the 12 subsections of the Review Checklist described earlier.
Evaluation survey At the time of receiving the final collated review for their protocol, a request was made to all chief investigators (i.e. those submitting their draft protocol for review) to provide feedback. They were asked to state whether or not they agreed with the issues raised at review and provide confirmation of changes made to their protocol as a result. A service evaluation audit was conducted to check chief investigator response rates and acceptance rates (% of remarks raised at review that were accepted and reflected in changes to the protocol) of proposed changes. This was done for all protocols reviewed in this 6-year period.
Statistical analysis Descriptive statistics were performed to present trial characteristics, frequency and type of remarks and response and acceptance rates. A response rate of 60% is considered as an acceptable level of response rate to surveys.14 All analyses were conducted using MicrosoftÒ Excel 2011 (Microsoft, Inc., Redmond, WA) and IBM SPSSÒ v.19 (SPSS, Inc., Chicago, IL) software.
Results Protocols reviewed by CPAS and findings Trial characteristics of the 176 protocols that were reviewed are described in Table 1. The median number of protocols per year reviewed was 27 (range: 25–42) and appears to be stable over the years. Of these, 4%
Debruyne et al.
5
Table 1. Characteristics of the clinical trials reviewed between 2008 and 2013. Characteristic
Year of review 2008 2009 2010 2011 2012 2013 Type of trial Phase I Phase I/II Phase II Phase II/III Phase III Phase IV Other Clinical specialty group Biomarkers and imaging Bladder (including penile) Brain Breast Children’s cancer and leukaemia Colorectal Gynaecological Haematological oncology Head and neck Lung Lymphoma Melanoma Palliative and supportive care Primary care Prostate Psychosocial oncology Renal (including adrenal) Sarcoma Teenage and young adults Testis Upper gastrointestinal (including pancreas and liver) Combined Type of investigational therapy Cytotoxic chemotherapy Hormonal therapy Molecular targeted therapy Immunotherapy Combination of drugs Combination with radiotherapy Other
Studies (N = 176) No.
%
29 26 28 42 25 26
16 15 16 24 14 15
7 19 90 13 44 1 2
4 11 51 7 25 1 1
0 9 11 16 3 14 16 26 6 10 13 7 1 0 8 0 7 8 0 2 10
0 5 6 9 2 9 9 15 3 6 7 4 1 0 5 0 4 5 0 1 6
9
5
43 3 42 1 63 15 9
24 2 24 1 36 9 5
were phase I trials, 51% were phase II trials, 25% were phase III trials, 1% were phase IV trials and 18% were combined phase I/II or II/III trials. The average time between submission of the protocol to CPAS and the return of the collated review was 35 days. This figure fluctuates between 21 days (and in one or two exceptional cases an excess of 80 days), depending on how busy CPAS service becomes at any one time. The disease categories subdivided according to the respective clinical specialty groups are listed in Table 1.
The included experimental treatment modalities concerned mainly cytotoxic chemotherapy (24%), molecular targeted therapy (24%) or a combination of drugs (36%). About 9% concerned studies of drug combinations with radiotherapy, 2% anti-hormonal treatment and 1% immunotherapy. The review findings according to the 12 categories in the Protocol Review Checklist and the relative frequencies that each issue has arisen in the respective protocols are summarised in Figure 2(a) and (b). The median number of remarks per protocol was 26, of which 20 were deemed clinically relevant. In our experience, the nature of the comments raised by reviewers fell into two broad categories: missing information and insufficient clarity of the information or guidance provided. The majority of clinically relevant remarks concerned the regimen (median [Q1, Q3]; 3 [1, 6]), support medication (2 [1, 4]), monitoring (2 [1, 3]) and drug supply aspects (2 [1, 4]) and were the same over the years (Figure 2(a)). Some typical examples are listed in Table 2.
Service evaluation survey A service evaluation survey was systematically sent to the respective chief investigators, of which 62% responded. All responses were positive (qualitative responses; data not shown), which were reflected in a median overall acceptance rate of 89% of the CPAS proposed protocol changes. Response rates and acceptance rates seemed to remain stable over the respective years (data not shown).
Discussion Cancer prevalence is high and will only increase in the near future. The quality of research protocols, in particular the pharmacy-related content, is of utmost importance to ensure cost-efficient, timely and safe research. To the best of our knowledge, an initiative like CPAS is unique, and such activities that aim to support chief investigators improve the pharmacy-related quality of cancer clinical trial protocols have not been published earlier. This article describes the development of CPAS and reviews its activities from 2008 until 2013. The aim of CPAS is to raise awareness of the type, frequency and consequences of research protocol inadequacies and to provide support to investigators either directly or in the form of guidance documents to assist the development of high-quality clinical trial protocols. CPAS was established in response to a general perception by our research community that pharmacy departments were a barrier to starting and running clinical trials especially those involving complex chemotherapy regimens. It was not the intention of pharmacy departments to hinder research, and within our pharmacy community, it was recognised that the workload generated by inadequate pharmacy information
6
Clinical Trials
Figure 2. (a) Number of remarks per review category per protocol, presented as boxplots, graphically displaying median, interquartile range and minimum and maximum data values and (b) relative frequency of remarks per review category presented as a pie chart.
within protocols often caused resource issues and subsequent delays to research.6 A number of issues were raised that hindered trial implementation at the local level and which caused delays and problems in areas such as dose adjustments, dose capping, missing pharmacy information, supply of drugs and safe administration of chemotherapy.6 Local practice often differs between hospitals causing inconsistent or wrong interpretation, which can negatively affect trial conduct and consequently data accuracy. Moreover, any missing or unclear protocol information has the potential to adversely affect patient safety. The remit of CPAS is, therefore, to resolve these issues at the draft protocol stage and achieve consistency across the clinical trial portfolio hosted by the NIHR CRN. Qualitative evaluation of the CPAS review process, through the service evaluation survey, presented positive feedback. Indeed, the majority of investigators provided a written response to the final collated review,
reporting that it was a helpful and constructive process that, in their opinion, reduced the number of protocol amendments that were required during the trial. The effectiveness of the service was also demonstrated by the remarkably high acceptance rate of the remarks raised by CPAS. The chief investigators accepted almost 9 out of 10 proposed amendments, which underscore the relevance of CPAS. While CPAS activities have been appreciated by the chief investigators of cancer clinical drug trials, the cost–benefit of this timeextensive review process is not verifiable at present. Future research could, therefore, prospectively examine whether an upfront pharmacy review process reduces the number of required pharmacy-related amendments, treatment-related protocol violations or the percentage of treatment-related hospitalisations or deaths. Moreover, future work to inspect the characteristics of clinical drug trials where the chief investigators did not respond to review comments or incorporate suggested
Debruyne et al.
7
Table 2. Examples of common relevant findings. Regimen No information on which drugs are IMPs and therefore which need accountability Information missing on infusion times, stability of a product, diluents, use of non-PVC infusion bags and giving sets Information missing on what to do if a patient vomits following a dose or misses a dose Use of brand names instead of generic Use of drug names that are not used in the United Kingdom, for example, acetaminophen instead of paracetamol Information copied from previous protocol resulting in incorrect information being stated Support medication No advice given for supportive medicines, for example, pre-meds, anti-emetics, hydration and so on 2 mg/L magnesium sulphate and 20 mmol post-hydration bags insisted on by protocol (2 mg/L = 0.008 mmol magnesium per litre) Dose calculation No information on frequency of re-calculation of BSA or formula to use, re-calculation of GFR and method of GFR calculation No reference to dose banding Dose banding table of chemotherapy with doses expected to be measured to two decimal places Monitoring Information missing on haematological and biochemical monitoring Different cut-offs specified in different areas of the protocol for discontinuing a drug due to renal impairment Screening investigations specified to be carried out within 14 days of treatment. Schedule of events table did not make this clear Drug administration Incorrect description of drug administration Different drug preparation available, incorrect choice made for route of administration required Proposed drug administration is not feasible in the specific study population GFR: glomerular filtration rate; PVC: poly(vinyl chloride); IMP: investigational medicinal product; BSA: body surface area.
changes could identify areas of further guidance or educational resource that CPAS could provide. Retrospective analysis did, at present, not reveal any differences in trial phase, clinical specialty group or type of investigational therapy between cooperative and non-cooperative chief investigators (data not shown). The latter might have been reluctant to wait for CPAS feedback since it is recognised that the turnaround time for CPAS review and final collation could be seen as an addition of a substantial amount of time to an already lengthy protocol development period. As such, CPAS suggests that draft protocols are submitted as soon as possible, and the review is conducted in parallel with other protocol development processes. Currently, the number of reviewers determined to scrutinise a draft protocol for CPAS is chosen to give a range of experience, specialities and views from different regions in the United Kingdom. It is possible that future refinements to the Review Checklist and elaboration on the detail of guidance documents already provided might reduce the number of reviewers required and/or the time period of the review process. Our systematic retrospective analysis indicated that a median of 26 remarks were suggested per protocol and that the majority of remarks addressed by the reviewers were deemed clinically relevant. To our knowledge, this is the first time that such information, based on the use of a customised Review Checklist, is available. A formal request for more information about the protocol review process was sent to several cancer cooperative networks, however, only reciprocated by the National Cancer Institute and the Swiss Group for Clinical Cancer Research. Industry-funded trials or
trials executed within a large cooperative group often provide in-house protocol review by scientific disease-specific committees, not always incorporating a pharmacist. However, there is only limited or no transparency regarding the protocol review process, and thus, it serves to no benefit to external (academic) investigators. Our results indicate that the most frequently observed inconsistencies concerned the drug regimen, support medication, monitoring and drug supply aspects, thus highlighting the importance of collaboration between the oncology physician and clinical pharmacist at the time of protocol design. CPAS protocol review included trial protocols of all phases, an extensive number of clinical specialty groups and a wide variety of cancer treatments. Results might, however, not be extrapolable to commercially funded trials. Future prospective research, addressing trials sponsored by industry and different funding agencies, might be useful to explore the potential for differences in protocol quality. Over the years that CPAS has been operating, the number of review remarks per protocol appears to have remained stable. This fact could indicate that CPAS is unlikely to become redundant in the future. On the other hand, it could also suggest that CPAS guidance documents and checklists are not well known by investigators and the CPAS advisory function not well utilised. There are many factors involved in determining why an organisation does not learn from previous experience, and these could be explored by CPAS to maximise their effectiveness. Sensitisation of other cooperative groups in oncology and other medical subspecialties might also be required to ensure a more
8 widespread implementation of CPAS knowledge in the future. Continuous service evaluation audits and evaluation of the activities will no doubt lead to further service improvements and adaption of CPAS tools. Moreover, with growth, the number of activities performed by CPAS may also further expand to include advice on practical clinical trial issues relevant to pharmacists, such as dose banding, chemotherapy stability and compatibility issues and the incidence and avoidance of chemotherapy medication errors. In conclusion, we have described the development and activities of CPAS. Systematic analysis of mandated reviews of pharmacy-related aspects of cancer clinical trial protocols proved to be useful and improved the quality of the clinical trials hosted by the NIHR CRN. Moreover, with the refinement of previously published CPAS guidance, development of new CPAS guidance and lessons learned from the review process itself, we hope that this article will lead to a more widespread implementation of our knowledge by other academic groups and better, safer clinical research. Acknowledgements We would like to acknowledge the voluntary service of CPAS committee and panel members, the cooperation of clinical trial units and chief investigators managing the academic studies, staff across the Clinical Research Networks, the Clinical Studies’ Groups, Department of Health (England), the Welsh Assembly Government, the Scottish Government, the Northern Ireland Department of Health and Social Services, Cancer Research UK, Medical Research Council and members of the National Cancer Research Institute. We would also like to thank Mrs Liz Gardner for her careful reading of this manuscript.
Declaration of conflicting interests The authors have declared no conflicts of interest.
Funding Currently, CPAS is funded by NIHR CRN Cancer (formerly known as the National Cancer Research Network, or NCRN). The national pharmacy advisor post is also funded by NIHR CRN Cancer, hosted by the University of Leeds.
References 1. Ferlay J, Soerjomataram II, Dikshit R, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer 2015; 136: E359–E386. 2. Mistry M, Parkin DM, Ahmad AS, et al. Cancer incidence in the United Kingdom: projections to the year 2030. Br J Cancer 2011; 105: 1795–1803.
Clinical Trials 3. Cancer Research UK. Clinical trials and research, http:// www.cancerresearchuk.org (accessed September 2014). 4. Debruyne PR, Knott VE and Pattison NA. A call for rigorous research and transparent study reporting. Eur J Cancer Care 2013; 22: 421–422. 5. Chan AW, Tetzlaff JM, Altman DG, et al. SPIRIT 2013 statement: defining standard protocol items for clinical trials. Ann Int Med 2013; 158: 200–207. 6. Blake D, Root T, Summerhayes M, et al. The NHS cancer plan and the pharmacy contribution to cancer care, 2001, http://www.bopawebsite.org/contentimages/publi cations/OncologyPharmacy.pdf 7. Stead M, Cameron D, Lester N, et al. Strengthening clinical cancer research in the United Kingdom. Br J Cancer 2011; 104: 1529–1534. 8. NIHR CRN Cancer. National Cancer Research Network (NCRN), http://www.crn.nihr.ac.uk/cancer/about-cancerresearch/chemotherapy-and-pharmacy-advisory-service (2013, accessed March 2014). 9. NIHR CRN Cancer. Chemotherapy and Pharmacy Advisory Service (CPAS) on behalf of the National Institute for Health Research Clinical Research Network (NIHR CRN) Cancer. Protocol review guidelines, http:// www.crn.nihr.ac.uk/cancer/about-cancer-research/chemotherapy-and-pharmacy-advisory-service (2013, accessed March 2014). 10. NIHR CRN Cancer. Chemotherapy and Pharmacy Advisory Service (CPAS) on behalf of the National Institute for Health Research Clinical Research Network (NIHR CRN) Cancer. Guidance on the drug-related content of clinical trial protocols, http://www.crn.nihr.ac.uk/cancer/ about-cancer-research/chemotherapy-and-pharmacyadvisory-service (2008, accessed March 2014). 11. NIHR CRN Cancer. Chemotherapy and Pharmacy Advisory Service (CPAS) on behalf of the National Institute for Health Research Clinical Research Network (NIHR CRN) Cancer. Guidance on the content of a pharmacy manual to support clinical trial protocols, http://www.crn.nihr.ac.uk/cancer/about-cancer-research/ chemotherapy-and-pharmacy-advisory-service (2012, accessed March 2014). 12. NIHR CRN Cancer. Chemotherapy and Pharmacy Advisory Service (CPAS) on behalf of the National Institute for Health Research Clinical Research Network (NIHR CRN) Cancer. Investigational Medicinal Product statement, http://www.crn.nihr.ac.uk/cancer/about-cancerresearch/chemotherapy-and-pharmacy-advisory-service (2011, accessed March 2014). 13. NIHR CRN Cancer. Chemotherapy and Pharmacy Advisory Service (CPAS) on behalf of the National Institute for Health Research Clinical Research Network (NIHR CRN) Cancer. List of protocols reviewed by CPAS to date, http://www.crn.nihr.ac.uk/cancer/about-cancerresearch/chemotherapy-and-pharmacy-advisory-service (2014, accessed March 2014). 14. Johnson TP and Wislar JS. Response rates and nonresponse errors in surveys. JAMA 2012; 307: 1805–1806.
